DESCRIPTION (adapted from applicant's application): Inhaled anesthetics are the most toxic and least understood drugs that physicians currently use. Improvements can only occur via a detailed understanding of their binding interactions, sites and targets. In this subproject of the program, we will continue to relate structural features of candidate protein targets with the stoichiometry and energetics of inhaled anesthetic binding using both isothermal titration calorimetry and amide hydrogen exchange (aim 1). Each target is structurally defined, so that features, such as cavity number, shape, volume, and polarity can be correlated with binding energetics. These features will be parameterized into a predictive, anesthetic binding algorithm, using the existing high-resolution apoferritin and serum albumin complexes for training, to mine further potential targets from the Protein Data Bank and the Protein Structure Initiative. In aim 2, proteins from aim 1 that show specific binding will be co-crystallized with various inhaled anesthetics to gain further structural detail to be incorporated into the binding algorithm, to use in correlative analyses of affinity, and to probe for the structural consequences of binding. In addition to apoferritin and serum albumin, 4 proteins were identified in the last cycle that will be examined initially. Finally, in aim 3, the powerful method of anesthetic photolabeling will be further refined in two ways. First, combining photolabeling with crystallography will allow validation of revealed location, and an estimate of selectivity and mechanism of incorporation. Continued development of diazirine and diazo volatile compounds will occur, and emphasis will be placed on adduction detection methods. Important interactions will each of the other 4 projects exist;these aims will contribute to the overall program goals by revealing the basis for affinity and stoichiometry- the energetic driver for altering protein conformation, and by revealing new and unanticipated inhaled anesthetic protein targets - either from the mining of existing databases, or through identification via photolabeling.